A conventional mobile device often includes an accelerometer that must be calibrated. Accelerometer calibration can be performed during manufacture of the mobile device or during use of the mobile device by an end user.
When calibrating the accelerometer during fabrication of the mobile device, the accelerometer is placed in a test rig, subjected to a known acceleration, and the accelerometer's output signal is measured. This test provides an accurate measurement of an error in the accelerometer's output signal. However, factory calibration traditionally requires expensive specialized hardware and software. The time required to perform the test increases production time, which lowers the rate of production, increases costs, and decreases profits. Thus, it is desirable to eliminate factory calibration of accelerometers in mobile devices.
User calibration of the accelerometer is a classic alternative to factory calibration. In one type of user calibration process, the user intentionally moves the mobile device in a particular complex sequence of movements. This process burdens the user, and can led to incorrect calibration when the user errs while performing the sequence. In another type of conventional mobile device, an autocalibration sequence constantly runs, with hope that the user will perhaps perform the complex sequence of movements that the device requires for autocalibration. This on-going autocalibration process continuously wastes processor time, wastes battery energy, and successful calibration occurs erratically. In addition to these concerns, conventional processes use algorithms that are subject to time-and temperature-induced errors causing accelerometer bias.
Accordingly, there are long-felt industry needs for methods and apparatus that mitigate problems of conventional methods and apparatus, including apparatus and methods for accelerometer autocalibration in a mobile device.